{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# figs_reddit.ipynb\n",
    "\n",
    "Author: MB\n",
    "\n",
    "Date run: April 1, 2021\n",
    "\n",
    "Location: HPC\n",
    "\n",
    "The purpose of this script is to read in a dataset of retweets of Trump tweets and generate three figures\n",
    "* the cumulative number of Reddit posts over time\n",
    "* the average number of page subscribers per tweet post on Reddit\n",
    "\n",
    "Data in: \n",
    "* Decahose (filtered to just retweets of Trump tweets) -- 10% of all tweets\n",
    "* Dataset of interventions to tweets\n",
    "* labels for Trump tweets\n",
    "* reddit posts containing Trump tweets\n",
    "\n",
    "Data out: the aforementioned figures"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# imports \n",
    "import datetime\n",
    "import glob\n",
    "import json\n",
    "import os\n",
    "import warnings\n",
    "warnings.filterwarnings(action='ignore')\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "from tqdm import tqdm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "def fill_frame(frame, tweet_id):\n",
    "    '''\n",
    "    this function forward fills metrics to fill in missing minutes where there is not a record of the number of retweets\n",
    "    it then returns the filled dataframe\n",
    "    \n",
    "    :pd.DataFrame frame: a dataframe of tweets with the number of seconds since posting `seconds` and number of retweets at that timestamp `count`\n",
    "    '''\n",
    "    # sort dataframe in ascending order by seconds\n",
    "    frame = frame.sort_values('seconds')\n",
    "    \n",
    "    # append the missing seconds to the dataframe with an empty count value\n",
    "    seconds_we_have = frame['seconds'].unique()\n",
    "    frame['count'] = range(len(frame)) \n",
    "    frame['count'] = frame['count'] + 1\n",
    "    frame = frame.append([{'seconds':i} for i in range(0, 86401) if i not in seconds_we_have]).sort_values('seconds')\n",
    "    \n",
    "    # forward fill the dataframe (carry values up to the next non-null value)\n",
    "    frame['count'] = frame['count'].fillna(method='ffill')\n",
    "    frame['count'] = frame['count'].fillna(method='bfill')\n",
    "    frame['tweet_id'] = tweet_id\n",
    "    \n",
    "    return frame\n",
    "\n",
    "def convert_datetime(time):\n",
    "    '''\n",
    "    converts twitter timestamp string into a datetime object\n",
    "    \n",
    "    :str time: the twitter-style timestamp for tweet creation\n",
    "    :returns date: a datetime.datetime object corresponding to the time input\n",
    "    '''\n",
    "    if isinstance(time, str):\n",
    "        return datetime.datetime.strptime(time, '%a %b %d %H:%M:%S +0000 %Y')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# set colors\n",
    "no_intervention_color = '#16697a'\n",
    "soft_intervention_color = '#ffa62b'\n",
    "hard_intervention_color = '#721121'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# read in the intervention dataset\n",
    "interventions = pd.read_json('../data/interventions.json', orient='records', lines=True, dtype={'id_str':str})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# read in the labels and unique trump tweets \n",
    "labels = pd.read_csv('../data/labels.csv')\n",
    "trump_tweets = pd.read_csv('../data/trump_tweets.csv', dtype={'retweeted__id':str})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# copy `Final` data to `political` because I like that better\n",
    "trump_tweets['political'] = labels['Final']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# filter Trump tweets only to tweets labelled as political\n",
    "political_tweets = trump_tweets[trump_tweets['political'] == 'political']['retweeted__id'].tolist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# filter interventions to Trump tweets that are political\n",
    "interventions = interventions[interventions['id_str'].isin(political_tweets)].drop_duplicates('id_str').copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# get tweet IDs by intervention type\n",
    "no_intervention = interventions[(interventions['hard_intervention'] == False) & (interventions['soft_intervention'] == False)]['id_str'].tolist()\n",
    "soft_intervention = interventions[(interventions['hard_intervention'] == False) & (interventions['soft_intervention'] == True)]['id_str'].tolist()\n",
    "hard_intervention = interventions[(interventions['hard_intervention'] == True)]['id_str'].tolist()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Reddit Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "reddit_posts = pd.read_json('../data/reddit_posts.json', orient='records', lines=True, dtype={'tweet_id':str})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# convert utc timestamp to datetime\n",
    "reddit_posts['date'] = reddit_posts['created_utc'].apply(datetime.datetime.utcfromtimestamp).dt.tz_localize('UTC')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# filter to tweets we have intervention labels for\n",
    "reddit_posts = reddit_posts[reddit_posts['tweet_id'].isin(interventions['id_str'])].copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# join reddit post with twitter timestamp data \n",
    "reddit_timestamps = pd.merge(reddit_posts[['tweet_id', 'date']], interventions[['created_at', 'id_str']], \n",
    "                         left_on='tweet_id', right_on='id_str')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "# filter to reddit posts made within 24 hours of twitter post\n",
    "reddit_timestamps = reddit_timestamps[(reddit_timestamps['date'] - reddit_timestamps['created_at'] > datetime.timedelta(minutes=0))\n",
    "                              & (reddit_timestamps['date'] - reddit_timestamps['created_at'] <= datetime.timedelta(hours=24))\n",
    "                             ]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "# calculate time difference \n",
    "reddit_timestamps['time_diff'] = (reddit_timestamps['date'] - reddit_timestamps['created_at']).values / np.timedelta64(1, 's')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "# calculate hours, minutes, and seconds since tweet creation\n",
    "reddit_timestamps['seconds'] = reddit_timestamps['time_diff']\n",
    "reddit_timestamps['minutes'] = reddit_timestamps['seconds'] / 60\n",
    "reddit_timestamps['hours'] = reddit_timestamps['minutes'] / 60"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "# forward fill null minutes\n",
    "filled_frames = pd.concat([fill_frame(frame, _) for _, frame in reddit_timestamps.groupby('tweet_id')])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "# filter by intervention type\n",
    "hard = filled_frames[filled_frames['tweet_id'].isin(hard_intervention)].copy()\n",
    "soft = filled_frames[filled_frames['tweet_id'].isin(soft_intervention)].copy()\n",
    "none = filled_frames[filled_frames['tweet_id'].isin(no_intervention)].copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "# get rolling average of cumulative posts for each intervention type\n",
    "hard_plot = hard.groupby('seconds')['count'].mean().rolling(window=3600, min_periods=20).mean()\n",
    "soft_plot = soft.groupby('seconds')['count'].mean().rolling(window=3600, min_periods=20).mean()\n",
    "none_plot = none.groupby('seconds')['count'].mean().rolling(window=3600, min_periods=20).mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot growth of reddit posts by intervention type since time that the tweet was created \n",
    "plt.figure(figsize=(15,10))\n",
    "plt.plot(hard_plot, label='Hard Intervention', color=hard_intervention_color, linewidth=4)\n",
    "plt.plot(soft_plot, label='Soft Intervention', color=soft_intervention_color, linewidth=4)\n",
    "plt.plot(none_plot, label='No Intervention', color=no_intervention_color, linewidth=4)\n",
    "\n",
    "plt.legend(fontsize=14)\n",
    "locs = [i*3600 for i in range(25)]\n",
    "plt.xticks(locs, range(25), rotation=45, fontsize=14)\n",
    "plt.yticks(fontsize=14)\n",
    "plt.legend(fontsize=14)\n",
    "plt.xlabel('Time since tweet publication (hours)', fontsize=14)\n",
    "plt.ylabel('Number of Reddit Posts', fontsize=14)\n",
    "plt.title('Growth of Reddit Posts by Intervention', fontsize=14)\n",
    "plt.gca().spines['top'].set_visible(False)\n",
    "plt.gca().spines['right'].set_visible(False)\n",
    "plt.savefig('../plots/reddit_spread_of_tweets.png', dpi=600)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "# filter reddit posts by intervention type\n",
    "hard_ct = reddit_posts[reddit_posts['tweet_id'].isin(hard_intervention)].copy()\n",
    "soft_ct = reddit_posts[reddit_posts['tweet_id'].isin(soft_intervention)].copy()\n",
    "none_ct = reddit_posts[reddit_posts['tweet_id'].isin(no_intervention)].copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot average page subscribers per tweet by intervention type\n",
    "plt.figure(figsize=(8, 10))\n",
    "\n",
    "plt.bar(0, width=.5, height=hard_ct.groupby('tweet_id')['subreddit_subscribers'].sum().mean(), color=hard_intervention_color, label='Hard Intervention')\n",
    "plt.bar(1, width=.5, height=soft_ct.groupby('tweet_id')['subreddit_subscribers'].sum().mean(), color=soft_intervention_color, label='Soft Itervention')\n",
    "plt.bar(2, width=.5, height=none_ct.groupby('tweet_id')['subreddit_subscribers'].sum().mean(), color=no_intervention_color, label='No Intervention')\n",
    "plt.legend(fontsize=14)\n",
    "plt.xticks([])\n",
    "plt.yticks(fontsize=14)\n",
    "plt.legend(fontsize=14)\n",
    "plt.xlabel('Intervention Type', fontsize=14)\n",
    "plt.ylabel('Average Total Page Subscribers', fontsize=14)\n",
    "plt.title('Average Total Page Subscribers per Tweet by Intervention Type (Reddit)', fontsize=14)\n",
    "plt.gca().spines['top'].set_visible(False)\n",
    "plt.gca().spines['right'].set_visible(False)\n",
    "plt.savefig('../plots/reddit_avg_exposure.png', dpi=600)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}
